In recent years, the commercial market for memory has gradually shift from that for personal, desk top computers to consumer devices such as handheld or portable music players, communication devices, computing devices, and combinations of these features in one device. Due to the smaller and smaller size of these portable devices, memory with higher density and speed, lower power consumption, and small size is in high demand.
Extensive development has been directed to non-volatile memory devices including FLASH, polymer random access memory (RAM), magnetic RAM (MRAM), ferroelectric RAM (FeRAM) and resistance RAM (RRAM). Magnetic shift registers have been proposed as an alternative to these RAM devices, as magnetic shift registers can store and access a large number of data bits using just a few logic elements and a low cost.
To facilitate the storage of data in magnetic shift registers, embodiments of shift registers that include indentations or protrusions into the width of the track to support a domain wall have been proposed. The indentations and protrusions, however, are difficult to form in the shift registers and are hard to control. What is needed is a system to provide better stability to domain walls and a better process to provide the stability.